ABSTRACT

Mammalian gut microbes colonize the intestinal tract of their host and adapt to establish a microbial ecosystem. The host diet changes the nutrient profile of the intestine and has a high impact on microbiota composition. Genetic mutations in Escherichia coli, a prevalent species in the human gut, allow for adaptation to the mammalian intestine, as reported in previous studies. However, the extent of colonization fitness in the intestine elevated by genetic mutation and the effects of diet change on these mutations in E. coli are still poorly known. Here, we show that notable mutations in sugar metabolism-related genes (gatC, araC, and malI) were detected in the E. coli K-12 genome just 2 weeks after colonization in the germ-free mouse intestine. In addition to elevated fitness by deletion of gatC, as previously reported, deletion of araC and malI also elevated E. coli fitness in the murine intestine in a host diet-dependent manner. In vitro cultures of medium containing nutrients abundant in the intestine (e.g., galactose, N-acetylglucosamine, and asparagine) also showed increased E. coli fitness after deletion of the genes-of-interest associated with their metabolism. Furthermore, the host diet was found to influence the developmental trajectory of gene mutations in E. coli. Taken together, we suggest that genetic mutations in E. coli are selected in response to the intestinal environment, which facilitates efficient utilization of nutrients abundant in the intestine under laboratory conditions. Our study offers some insight into the possible adaptation mechanisms of gut microbes.

IMPORTANCE

The gut microbiota is closely associated with human health and is greatly impacted by the host diet. Bacteria such as Escherichia coli live in the gut all throughout the life of a human host and adapt to the intestinal environment. Adaptive mutations in E. coli are reported to enhance fitness in the mammalian intestine, but to what extent is still poorly known. It is also unknown whether the host diet affects what genes are mutated and to what extent fitness is affected. This study suggests that genetic mutations in the E. coli K-12 strain are selected in response to the intestinal environment and facilitate efficient utilization of abundant nutrients in the germ-free mouse intestine. Our study provides a better understanding of these intestinal adaptation mechanisms of gut microbes.

INTRODUCTION

The human gut harbors as many gut microbes as host cells (1). The total number of genes carried by the gut microbiota is estimated to be greater than the number of host genes (2, 3). The gut microbiota creates a complex ecosystem through interactions between bacteria and host cells. Although the host genetic background influences the composition of the gut microbiota (4), cohort data indicate that environmental factors, including diet, have a significantly greater impact (5). Microbiota-accessible carbohydrates (MACs) are found in host dietary fibers, and they are the primary nutrients utilized by the gut microbiota. MACs greatly influence the gut microbiota composition (68). Long-term changes from high-MAC diets to low-MAC diets result in the irreversible loss of certain species (9). Dietary composition exerts selective pressure for not only species but also variants and substrains within the same species (10, 11), suggesting that genetic mutations in the gut bacterial genome occur during adaptation to the gut environment. However, these studies have mainly focused on the relative abundance of species or bacterial genes and information on the adaptation process itself, including the selection of advantageous mutations, is lacking.
Escherichia coli, which lives in the human gut throughout the life of a human host (1214), is widely used for experimental adaptation studies due to the ease of genetic engineering. Recently, experiments using E. coli-colonized mice have uncovered several adaptive mutations in the E. coli genome (1518). The functions of the mutated genes identified were mainly involved in bacterial metabolism, for example, gat genes involved in galactitol metabolism (16), lrp encoding a global transcriptional regulator of amino acid metabolism (17), and dgoR encoding a repressor of galactonate metabolism genes (18). Furthermore, mutated genes were altered by co-colonization with Blautia coccoides through modulation of the gut metabolome (17). These findings suggest that the genetic mutation of E. coli reflects the intestinal metabolome environment, where various metabolites are present. In addition to the existence of other bacterial species, the host diet can also alter the gut environment. Indeed, the genetic signature of Bacteroides thetaiotaomicron during colonization of the murine intestine is diet specific, and daily switching of diet increases genetic diversity (19). Metagenomic fragments collected from mice inoculated with probiotic E. coli Nissle 1917 showed that certain gene functions were enriched depending on host dietary condition (15), suggesting that diet affects genetic mutation within the E. coli genome during adaptation. However, the extent to which dietary changes affect the fitness of such mutations in E. coli and whether the changes alter the mutational landscape of E. coli colonization in the murine intestine are still poorly known. Therefore, we focused on genetic mutations during intestinal colonization in this study.
To investigate the relationship between the host’s gut environment and mutations, we used an E. coli hyper-mutator strain (ΔmutL and ΔmutS), which is derived from E. coli K-12 strain, and germ-free (GF) mice. Hyper-mutator strains are more likely to accumulate mutations in their genomes, and GF mice enable us to clearly detect mutations of these strains within the mammalian intestine without the interference of other variables such as other resident bacteria. This framework streamlined our investigation on the evolution of highly adaptive E. coli strains within the host gut environment. Through the study results, we were able to confirm that the deletion of several genes commonly mutated in independent experiments elevated the fitness of E. coli in the murine intestine, and we observed that the extent of fitness depended on the host diet. Furthermore, we found that in an in vitro setting, these gene deletions enhanced the ability to utilize specific nutrients that was abundant in the intestinal lumen of mice depending on their diet.

RESULTS

Screening of mutated genes in E. coli during murine intestinal colonization

We investigated the adaptation of Escherichia coli to the mouse gut by colonizing GF mice with an Escherichia coli K-12 hyper-mutator strain (ΔmutL or ΔmutS) (20, 21). The mut genes encode a mismatch repair (MMR) protein that normally prevents genetic aberrations in the genome. The mutation frequencies in the ΔmutL and ΔmutS strains are more than 100-fold higher than in the wild-type (22, 23). In vitro studies have shown that mutator-specific mutations are very few relative to all mutations (24). We inoculated the hyper-mutator strains into three groups of GF mice to detect commonly mutated genes in independent experiments (Fig. 1A). Rapid gene mutations were observed, and the mutation trajectories of E. coli within the same experiment tended to be similar (Fig. 1B).
Fig 1
Fig 1 Mutated genes in hyper-mutator E. coli during adaptation to the murine intestinal environment. (A) Schematic representation of the mutated gene screening experiment. (B) Mutated genes in E. coli during intestinal colonization in three independent experiments. Each column on the heatmap indicates a sampling timepoint of a mouse, whereas each row indicates a mutated gene that was estimated to have “HIGH” or ”MODERATE” effect on protein function by SnpEff (25). Estimates derived from mutations at different positions within the same gene were consolidated. The dendrogram on the left was calculated by complete clustering of Euclidean distance. The boxes on the bottom represent the sampling points of feces, experiments, and mouse replicates. The first experiment using ΔmutL: n = 3; second: n = 2; and ΔmutS: n = 3. Heatmap color indicates the frequency of the mutation in each mouse. (C) Venn diagram of commonly mutated genes in two or more mice in each experiment on day 84.
We identified the mutations that likely affected protein function (e.g., non-synonymous coding, frameshift, and stop gain mutations; see Materials and Methods) from the day 84 samples and found that nine gene mutations were commonly detected in two or more experiments (Fig. 1C). Mutations in gatC, araC, and malI genes were observed in most mice at day 14, and all mice at day 84 (Fig. S1). Frameshift and stop codon acquisition mutations were observed in these three genes (Table 1), suggesting that their loss of function contributed to intestinal colonization. These genes encode a galactitol transporter (26), transcriptional factors for the arabinose operon (27, 28), and that for the maltose operon (29), respectively (Table 1). Mutations in the gat operon—which is responsible for galactitol metabolism—in E. coli strain K-12 have been reported to occur in the mouse intestine in previous studies (16, 17, 30). The gat operon is constitutively expressed in E. coli K-12 because the gat operon repressor gene, gatR, is dysfunctional due to an insertion sequence (IS) (16, 31). In variants of this strain in which the gatR pseudogene has been replaced by a functional version, inactivation of gatR by IS has been observed to occur during intestinal colonization in mice (31). However, the growth of wild-type E. coli K-12, in which the expression of gat operon genes was enhanced, was inhibited by galactitol in vitro, while the gat gene-mutated strain was not (16). Because the hyper-mutator strains used in our study are derived from E. coli K-12, which possess a dysfunctional gatR gene due to the insertion sequence, we considered that the accumulation of the gatC mutation might be a consequence of constitutive gat operon expression in E. coli K-12. Thus, we shifted our focus on the other gene mutations in accordance with our original goal of investigating the adaptation to the intestinal environment. In regard to malI, a previous study observed malI mutations in some E. coli K-12-inoculated mice. However, mutations in this gene were not prevalently detected in all mice used in the study (32).
TABLE 1
TABLE 1 Commonly mutated genes in three experiments on day 84
GeneFunctionPositionRef.Alt.Mutation frequency (%)aMutation type
araCArabinose operon regulatory protein70424AG66.40Non-synonymous coding
70484AAT24.92Frame shift
70517GA35.06Non-synonymous coding
70547AGA22.21Frame shift
70698GA25.00Stop gained
dgsADNA-binding transcriptional repressor Mlc1665754GA36.02Stop gained
1665849TC56.61Non-synonymous coding
1665912GGC49.54Frame shift
1666162GGT44.79Frame shift
1666357GCG28.49Frame shift
fruK1-Phosphofructokinase2259847GA47.99Non-synonymous coding
2260075CT57.38Non-synonymous coding
gatCPTS system galactitol-specific EIIC component2171258AG23.12Non-synonymous coding
2171319GA62.73Stop gained
2171472GA64.39Stop gained
2171604GCG56.58Frame shift
2171775CTC26.86Frame shift
malIMaltose regulon regulatory protein1696432GCG26.46Frame shift
1696596GGC56.32Frame shift
1696908CCA21.28Frame shift
1696908CAC61.33Frame shift
1697167GA22.75Non-synonymous coding
melBMelibiose:H+/Na+/Li+symporter4341572AG23.49Non-synonymous coding
4341740GA31.15Non-synonymous coding
4342227AG26.33Non-synonymous coding
4342446TC52.61Non-synonymous coding
4342508AG26.17Non-synonymous coding
melRMelibiose operon regulatory protein4338978AG27.68Non-synonymous coding
4339297CT50.62Non-synonymous coding
4339530CT21.30Non-synonymous coding
ompCOuter membrane porin C2310395TC53.25Non-synonymous coding
ygjIInner membrane transporter3224549CCT28.64Frame shift
3224687CCG52.62Frame shift
3224687CGC21.39Frame shift
a
Mean frequency of mutation in E. coli within all mice that harbored this mutation on day 84.

The mutated genes contribute to increased colonization fitness in the mouse intestine

To investigate the contributions of these gene mutations on fitness in the mouse intestine, we constructed multiple-mutant strains. Here, we focused on the araC and malI genes, whose mutations were commonly detected in all mice. Competitive fitness of araC and malI mutants was compared with ΔgatC, as the gat gene mutation is quickly introduced during murine intestinal colonization, is eventually acquired at a consistently high frequency, and appears to have the great impact on fitness (16). As such, we introduced this highly dominant mutation to our strains as a baseline to diminish its influence on our results and compare it with other mutations. Double mutants (ΔgatCΔaraC, ΔgatCΔmalI, and ΔgatCΔmelR) and a triple mutant (ΔgatCΔaraCΔmalI) were co-inoculated with a ΔgatC single-mutant in GF mice to compare their ability to establish intestinal colonization (Fig. 2A). We used ΔgatCΔmelR as the baseline for comparison of ΔgatCΔaraC and ΔgatCΔmalI since the mutation in melR was detected in only two groups during colonization of the hyper-mutator strain, which suggests that any fitness conferred by melR mutation is assumed to be less than that of araC or malI mutation. In addition, melR, which encodes a transcription factor for the melibiose operon, shares a common feature with araC and malI in that it also encodes a sugar metabolism-related transcription factor (Fig. 1C).
Fig 2
Fig 2 Deletion of mutated genes confers fitness to E. coli in the murine intestine. (A) Schematic representation of in vivo competition assay between ΔgatC and double- or triple-mutant strains. (B) In vivo competition assay between ΔgatC and multiple-mutant strains (n = 4). The log10 competition index (CFU of double- or triple-mutants divided by that of ΔgatC) is shown as the mean  ±  SEM. *P  <  0.05 (paired t test with Holm’s correction between ΔgatC and each double or triple mutant on day 14). (C) Principal component analysis (PCA) score plot of sugar concentration in the feces sampled on day 14 (left panel) and its loadings (right panel). In the PCA panel, each point represents a sample, and the color indicates inoculated strains. P value was calculated with PERMANOVA.
By fourteen days after inoculation, ΔgatCΔaraCΔmalI, ΔgatCΔaraC, and ΔgatCΔmalI significantly outcompeted ΔgatC by a >1,000-fold increase in CFU (Fig. 2B). ΔgatCΔmelR also outcompeted ΔgatC, but the CFU of ΔgatCΔmelR was just 1.5-fold greater than that of ΔgatC. These results suggest that the loss of function of araC and malI enhances colonization fitness in the murine intestine for E. coli. The competition index was highest in the order of ΔgatCΔaraCΔmalI, ΔgatCΔaraC, and ΔgatCΔmalI. This suggests that araC-deficiency confers a greater fitness increase to E. coli than malI deficiency, and deficiency in both genes additively contributes to colonization fitness. Given that the mutations in araC, gatC, and malI were detected in most of the mice on day 14 (Fig. S1), mutations resulting in highly elevated fitness would be observed earlier during colonization than those with lesser or negligible impact.
Since araC and malI are related to sugar utilization, we measured the amounts of sugars in the E. coli mutant-colonized mouse intestine to identify possible patterns that would implicate sugar metabolism with intestinal fitness. Galactitol was detected in mouse cecal contents, colon contents, and feces (Fig. S2A), suggesting that the occurrence of gatC mutations (Fig. 1C) was to avoid galactitol metabolism because galactitol metabolism is known to severely inhibit growth for unknown reasons (16). Principal component analysis (PCA) score plots of sugar profiles based on the fecal concentrations of 25 different sugars were distinct in each mouse group, implying that each gene mutation differentially influences the sugar utilization patterns of E. coli in the murine intestine (Fig. 2C). Therefore, we hypothesized that altering sugar metabolism in E. coli might be a key to obtaining greater fitness.

The mutated genes are optimized and selected for nutrient composition in the intestine, which changes with the host diet

Host diet, especially MACs, heavily influences the profile of intestinal carbohydrates, which the gut microbiota can utilize, and subsequently affects gut microbiota composition (69). We reasoned that if gene mutation-induced changes in sugar metabolism by E. coli enhance fitness in the intestine, dietary changes that affect sugar and other carbohydrate levels may lead to different results in competition between E. coli strains. We confirmed that intestinal sugar profiles in GF mice fed with different diets (high-MAC diet, used in mutated gene screening [Fig. 1A] and in vivo competition assays [Fig. 2A], and low-MAC diet [Table S1]) were significantly different (Fig. 3A). We co-inoculated two groups of GF mice that were fed these diets with ΔgatCΔaraCΔmalIΔmelR and ΔgatC. In both groups, ΔgatCΔaraCΔmalIΔmelR outcompeted ΔgatC. However, the competition index in the low-MAC diet group was 100,000 times smaller than that of the high-MAC diet group (Fig. 3B and C). The significant differences in sugar composition and quantity in feces suggest that certain carbohydrates might be involved in the fitness of the multiple-mutant strain. (Fig. 3A; Fig. S2B).
Fig 3
Fig 3 Trajectory of advantageous mutations depends on nutrient composition in the intestine. (A) Principal component analysis of the amounts of fecal sugars before inoculation (left panel) and its loading plot (right panel). In the PCA panel, each point represents a sample, and the color indicates the given diet. (B) Schematic representation of in vivo competition assay between ΔgatCΔaraCΔmalIΔmelR and ΔgatC and in different diet-fed mice. (C) In vivo competition assay between ΔgatCΔaraCΔmalIΔmelR and ΔgatC in different diet-fed mice. The log10 competition index (CFU of quadruple-mutant strain divided that of ΔgatC) is shown as the mean  ±  SEM (n = 4). The line pattern represents diet (high-MAC diet, solid; low-MAC diet, dashed). ***P  <  0.005 (two-way ANOVA). (D) Commonly mutated genes detected in more than half of high-MAC diet-fed mice (n = 8) or low-MAC diet-fed mice (n = 6) on day 84. Each column on the heatmap indicates a mouse individual, whereas each row indicates a mutated gene that was predicted to affect protein function. The frequencies of mutations at different positions within the same gene are consolidated. The boxes on the bottom denote the experiment IDs. Mutated genes detected in more than half of high-MAC diet-fed mice, low-MAC diet-fed mice, or both groups are written in green, purple, and black, respectively.
To determine whether host dietary changes affect the mutational trajectory, we inoculated ΔmutL in GF mice fed a low-MAC diet and monitored mutations in the E. coli genome. We observed that nine genes in the high-MAC diet group and 11 genes in the low-MAC diet group were mutated in more than half of the mice in each group (Fig. 3D; Table S2; see Materials and Methods). Although gatC, araC, and dgsA mutations were found in both groups, the other genes were found to be mutated in >50% of mice in their respective diet group and mutations were largely unique to host diet. One mouse in the low-MAC diet group (low-MAC 1) displayed a different pattern of mutated genes from the other mice. Although gatC mutation was not detected in this mouse, the gatB gene, which encodes for a galactitol transporter subunit along with gatC, was mutated. Therefore, the E. coli in the mouse seemed to avoid growth inhibition by galactitol metabolism by selecting for this mutation in a similar fashion as gatC-mutated strains in other mice. Although some mutated genes were detected in more than half of mice in both diets, the results suggested that the mutated gene was selected for nutrient composition in the intestinal lumen, which changes with the host diet.

The host diet alters the expression of genes in E. coli responsible for the metabolism of nutrients abundant in the intestine

The genes araC and malI, whose deletions conferred colonization fitness to E. coli in the high-MAC diet-fed mice, encode transcription factors. AraC activates the ara operon involved in arabinose metabolism in the presence of arabinose (27), the amount of which was higher in the high-MAC group than in the low-MAC group (Fig. S2B). Notably, the E. coli K-12 BW25113 strain we used cannot utilize arabinose as a carbon source because of the deletion of the araBAD gene. To determine whether arabinose is responsible for the advantage of araC deficiency, we cocultured ΔgatC and ΔgatCΔaraC in M9 + glucose minimal media with or without arabinose. As a result, although araC deficiency indeed increased fitness in the presence of arabinose (Fig. S3), the extent of increased fitness was smaller than that observed in vivo in terms of competition index (Fig. 2B). Considering how our E. coli strain should be unable to metabolize arabinose, araC deficiency in this strain may prevent transcription and translation of the ara genes in the presence of arabinose, which leads cells to use the energy to grow more efficiently. In addition, recent studies have shown that AraC regulates a wide range of gene expression (33), which suggests the involvement of other metabolic processes. MalI is a repressor of the malXY operon encoding maltose transporters (29), and the deletion of malI is expected to enhance maltose uptake and dissimilation. However, maltose was not detected in the feces of the high-MAC diet group prior to inoculation, suggesting that it was not readily available in the intestinal environment (Fig. S2B). Therefore, we speculated that colonization fitness in the high-MAC diet conferred by araC and malI deletion was due to other metabolites.
To elucidate the mechanisms of colonization fitness enhancement, we analyzed the influence of the deletion of araC and malI on gene expression in E. coli colonized in the intestine. We inoculated either ΔgatCΔaraCΔmalI or ΔgatC in GF mice fed a high-MAC or low-MAC diet and extracted total bacterial RNA. Differentially expressed genes (DEGs) between the strains differed according to the host diet (Fig. 4A). We hypothesized that DEGs detected only in high-MAC diet-fed mice (101 genes) might contribute to the enhanced fitness of ΔgatCΔaraCΔmalI in high-MAC diet-fed mice. The expression of galP, nagABE, and ydeM was upregulated in ΔgatCΔaraCΔmalI in the high-MAC diet-fed mice compared with ΔgatC (Fig. 4A; Table S3). galP encodes a galactose transporter (34), and nagABE is involved in the metabolism of N-acetylglucosamine (GlcNAc) (35, 36), a component of mucin. ydeM encodes an anaerobic sulfatase maturation enzyme (37) and is co-transcribed with ydeN, which encodes a sulfatase (38). ydeN is a direct target of AraC, and the ΔaraC strain highly expresses ydeMN (33). Although the detailed substrates of YdeN have not been reported, YdeN of E. coli K-12 strain showed 33% amino acid sequence similarity with BT4656, which encodes N-acetylglucosamine-6-O-sulfatase from Bacteroides thetaiotaomicron. Sulfatases are required for Bacteroides to access mucosal glycans in the mammalian intestine (39). Therefore, we speculate that ydeMN is related to mucin metabolism by E. coli. Among asparagine metabolism-related genes, the expression of the synthesis gene (40) (asnB, log2FC = −0.98, P = 0.001) was decreased, and the expression of the degradation gene (41) (ansB, log2FC = 1.01, P = 0.047) was increased (Fig. 4A). These results prompted us to test whether these metabolites are responsible for the colonization fitness of ΔgatCΔaraCΔmalI in the high-MAC diet. Indeed, the galactose, GlcNAc, and asparagine concentrations were higher in the feces of the high-MAC diet group (Fig. 4B; Fig. S4A).
Fig 4
Fig 4 AraC and malI deficiency enhances fitness in media containing the corresponding nutrients among those abundant in the intestines of high-MAC diet-fed mice. (A) A comparison of DEGs between ΔgatCΔaraCΔmalI and ΔgatC in high-MAC or low-MAC diet-fed mice. DEGs between the strains were assessed using DESeq2 (42). Venn diagram (left panel) shows the number of DEGs (FDR < 0.05 and log2 fold change > 1 or < −1) between ΔgatCΔaraCΔmalI and ΔgatC in high-MAC or low-MAC diet-fed mice (n = 3). The volcano plot (right panel) shows the P value and log2 fold changes of gene expression comparing ΔgatCΔaraCΔmalI and ΔgatC in high-MAC diet-fed mice. Transcripts with significant differential expression between the compared strains are highlighted according to designated color codes. (B) The amounts of metabolites in feces of high-MAC or low-MAC diet-fed mice before inoculation. ***P  <  0.005 (Welch’s t test with FDR correction). (C through G) In vitro competition assay between ΔgatC and double-mutant strains (n = 3) in M9 + 0.4% mucin (C), M9 + 0.4% galactose (D), M9 + 0.4% GlcNAc (E), M9 + 0.4% glucose + 0.4% asparagine (F), and M9 + 0.4% glucose (G). The log10 competition index (CFU of a double-mutant strain divided that of ΔgatC) is shown as the mean  ±  SEM. *P  <  0.05, **P  <  0.01, and ***P  <  0.005 (paired t test with Holm’s correction between ΔgatC and each double-mutant strain on day 14).

The advantage of araC and malI mutations is reproducible in vitro when their corresponding nutrients, abundant in mice fed a high-MAC diet, are utilized as a sole carbon source

To determine whether these metabolites are responsible for the fitness advantage associated with araC and malI deficiency, we co-cultured ΔgatCΔaraC, ΔgatCΔmalI, or ΔgatCΔmelR with ΔgatC in M9-based minimal media containing mucin, galactose, GlcNAc, or asparagine as a carbon source. Because the strains were not able to grow in M9 + asparagine medium, we used glucose + asparagine medium. Although the amount of glucose was also higher in the feces of the high-MAC diet group than in the low-MAC diet group, the average amount of glucose in the low-MAC diet group was 8.7 µmol/g, which was higher than most other sugars in the low-MAC diet group (Fig. 4B; Fig. S2B).
By 14 days after inoculation, ΔgatCΔaraC outcompeted ΔgatC in mucin medium (Fig. 4C). Mucin, the main component of mucus, is a source of nutrients for some intestinal microbes (e.g., B. thetaiotaomicron and Akkermansia muciniphila) (4345), and E. coli can also utilize mucin (46). Although the possibility that the commercial mucin (bound sialic acid 0.5 – 1.5%; free N-acetylneuraminic acid ≤ 0.2%) we used contained trace amounts of other growth-promoting compounds cannot be ruled out, the results suggest that the AraC regulon contains mucin degradation-related genes that allow for metabolism of mucin. ΔgatCΔmalI outcompeted ΔgatC in galactose, GlcNAc, glucose + asparagine, and glucose media (Fig. 4D through G). Although ΔgatCΔmalI outcompeted ΔgatC in glucose medium, the competition index was higher in glucose + asparagine medium, suggesting that asparagine conferred in vitro fitness to ΔgatCΔmalI (Fig. 4F and G). In summary, we predict that araC and malI deficiency enhances the metabolism of nutrients abundant in the mouse intestine, which results in increased intestinal colonization fitness. Interestingly, ΔgatCΔaraCΔmalI did not outcompete ΔgatC in the media containing mucin, galactose, and GlcNAc as a sole carbon source (Fig. S5A). However, the triple-mutant strain outcompeted ΔgatC in mucin + galactose medium. Furthermore, the competition index of ΔgatCΔaraCΔmalI was approximately threefold larger than that of ΔgatCΔaraC and ΔgatCΔmalI in the same medium (Fig. S5B).
GlcNAc, asparagine, and glucose were also present in the intestine of the low-MAC diet-fed mice (Fig. 4B), though at lower concentrations. Therefore, these metabolites might also contribute to the slight predominance of the quadruple-mutant strain in the low-MAC diet-fed mice (Fig. 3C).

DISCUSSION

In this study, rapid mutation of sugar metabolism-related genes was observed in E. coli mutator strains (Fig. 1BC; Fig. S1). Deletion of these mutated genes conferred colonization fitness to an extent that increased based on how many mice a given gene was mutated in (Fig. 2B). The host diet affected the selection of mutated genes and their associated fitness increase (Fig. 3C and D). The deletion of the two most commonly mutated genes, araC and malI, changed the expression of other metabolism-related genes responsible for the utilization of metabolites that were abundant in the intestinal environment in vivo (Fig. 4AB) and enhanced in vitro fitness in media containing these metabolites as carbon sources (Fig. 4C through G; Fig. S5).
The feature of mutated genes was consistent with previous studies on the point that metabolism-related genes were mutated (1619, 30), suggesting that genetic mutation of gut microbes strongly reflects the intestinal metabolome environment. The significant fitness impact of the deletion of araC and malI might be related to the fact that these genes encode transcription factors. In fact, AraC regulates the expression of a wide range of genes (33), which suggests the involvement of other metabolic processes. We expected araC deficiency to have a higher impact on fitness because in vivo competition assays showed that deletion of araC resulted in greater fitness in E. coli than malI (Fig. 2B). However, the enhanced fitness conferred by araC deficiency only seemed to be reflected in mucin and arabinose-supplemented media, and the extent of fitness tended to be lower than malI deficiency in our in vitro competition assay (Fig. 4C through G; Fig. S3). One possible implication of this is that other metabolites in the intestine contribute to the fitness-enhancing effects of araC inactivation. Any associations between malI and the metabolism of galactose, GlcNAc, and asparagine have not been reported. However, the DEGs between ΔgatCΔaraCΔmalI and ΔgatC in vivo and in vitro competition assays in our study suggest that MalI might regulate more metabolic processes (Fig. 4A and D through G). Some studies also reported that genetic mutations in E. coli during adaptation to the murine intestine promoted the metabolism of various metabolites through modulation of non-canonical regulatory targets. For example, a mutation in lacI, which encodes a transcription factor for the lactose operon, promoted raffinose metabolism, and a mutation in gntT, which encodes a gluconate transporter, promoted GlcNAc and mucin metabolism (15, 32). Interestingly, ΔgatCΔaraCΔmalI did not outcompete ΔgatC in mucin, galactose, and GlcNAc media in vitro but outcompeted in mucin + galactose medium (Fig. S5), suggesting that the gene mutations in araC and malI were selected simultaneously to optimize for the complex nutrient profile in the mammalian intestine.
The findings in this study are subject to two major limitations. The first is the use of GF mice. Indigenously colonized bacteria can affect the adaptation of invading bacteria through horizontal gene transfer (47, 48) and alteration of the metabolic environment of the gut (17). In particular, the existence of bacteria capable of metabolizing complex plant polysaccharides may greatly increase the carbon sources available to E coli. Therefore, mutation profiles and the extent of enhanced fitness by deletion of the commonly mutated genes, e.g., gatC, araC, and malI, might change with the existence of resident bacteria. It is important to acknowledge that this study focused on the relationships between mutated genes, their effected changes in gene expression, and metabolites in the intestine by excluding the variables introduced by other resident bacteria. The second is the use of E. coli K-12, which is a laboratory strain. E. coli K-12 is an attenuated strain, and it does not normally colonize the human intestinal tract (49). Both commensal and pathogenic E. coli strains inhabit the mammalian gut, and they exhibit diverse phenotypic and genotypic variants, including differences in carbon source utilization (50). These differences in E. coli strains should be considered when generalizing the results of this study.
Notwithstanding these limitations, we demonstrated that genetic mutations in E. coli are selected to optimize metabolism in response to the host diet-influenced gut environment. Recently, the gut microbiota has been targeted to prevent and treat diseases in humans. Some studies show attempts to manipulate it by using methods such as fecal microbiota transplantation (FMT) or probiotics (15, 51, 52). Our study showed that the metabolic profile, which can be altered by the host diet, has high impact on the colonization fitness of gut microbiota. This result suggests that the host diet might improve the above microbiota-targeting treatments. Overall, this study offers some insight into possible adaptation mechanism of the microbiota in the intestinal environment.

MATERIALS AND METHODS

Animals

The strain, genetic background, age, sex, replicates, and housing laboratory of the mice used in this study are listed in Table S4. The animals were bred and raised in axenic isolators under controlled light conditions (12-h light/12-h dark cycle). They were given sterilized water and fed CMF (Oriental Yeast Co. Ltd., Itabashi, Tokyo, Japan) as a high-MAC diet. Low-MAC diet-fed mice group changed their diet from CMF to AIN-93G (Oriental Yeast Co. Ltd., Itabashi, Tokyo, Japan) 1 week before the start of the experiment. Animal experiment procedures were approved by the Institutional Animal Care and Use Committee of the RIKEN Yokohama Branch and University of Tsukuba.

Bacterial strains

The Escherichia coli K-12 BW38029 hyper-mutator strain (ΔmutL) was provided by professor Hirotada Mori at the Nara Institute of Science and Technology (current: Guangdong Academy of Agricultural Sciences). The Escherichia coli K-12 BW25113 hyper-mutator strain (ΔmutS) (21) was provided by Dr. Ryuichi Koga at the National Institute of Advanced Industrial Science and Technology. The Escherichia coli K-12 BW25113 ΔgatC strain of the Keio collection (53) were purchased from the National BioResource Project (Mishima, Shizuoka, Japan). To construct multiple-mutant strains, we removed the ntpII gene, flanked by FLP recognition sites (FRT), from the single mutant genome by transformation of pFLP3. pFLP3 was a gift from Herbert Schweizer (Addgene plasmid #64946; https://www.addgene.org/64946/; RRID:Addgene_64946). We then replaced the target gene with a kanamycin resistance gene (neomycin phosphotransferase II: nptII) for subsequent gene deletion by λ-Red recombination. The FRT-flanked ntpII genes were amplified in each mutant genome using the primers listed in Table S5. The transformants were then screened on lysogeny broth (LB) agar plates supplemented with kanamycin (25 µg/mL). To construct ∆gatC[cmR], we introduced a chloramphenicol resistance gene (chloramphenicol acetyltransferase: cat) into the intS locus. Insertion of the antibiotic genes into the target region was confirmed by colony PCR.

Medium preparation

LB medium was prepared with 10 g/L tryptone (Becton, Dickinson and Company, Franklin Lakes, NJ, USA), 10 g/L NaCl (NACALAI TESQUE INC., Kyoto, Kyoto, Japan), and 5 g/L yeast extract (Becton, Dickinson and Company, Franklin Lakes, NJ, USA), adjusted to final volume with MilliQ water, and autoclaved. M9 + 0.4% glucose, galactose, N‐acetyl glucosamine (GlcNAc), and asparagine medium was made with M9 minimal salts, 5× (Becton, Dickinson and Company, Franklin Lakes, NJ, USA), following the manufacturer’s instructions. In brief, 56.4 g of the M9 minimal salts, 5× powder, was dissolved in 1 L of purified water and autoclaved at 121°C for 15 minutes. To prepare 1 L M9 + 0.4% carbon source media, the following components were mixed: 200 mL sterile M9 minimal salts, 5×; 778 mL sterile, purified water; 20 mL filter-sterilized solution of 20% D-(+)-glucose (NACALAI TESQUE Inc., Kyoto, Kyoto, Japan), D-(+)-galactose (FUJIFILM Wako Pure Chemical Corporation, Osaka, Osaka, Japan), N-acetyl-D-(+)-glucosamine (FUJIFILM Wako Pure Chemical Corporation, Osaka, Osaka, Japan), and L-asparagine monohydrate (Sigma-Aldrich Co. LLC, St. Louis, MO, USA); 2 mL sterile 1.0 M MgSO4 solution; and 0.1 mL sterile 1.0 M CaCl2. For the preparation of mucin-containing media (46), 0.2 M sodium dihydrogenphosphate buffer was prepared by dissolving 12 g sodium dihydrogenphosphate, Anhydrous (NACALAI TESQUE INC., Kyoto, Kyoto, Japan) in 500 mL MilliQ water, and 0.2 M disodium phosphate buffer was prepared by dissolving 14.2 g disodium hydrogen phosphate (NACALAI TESQUE INC., Kyoto, Kyoto, Japan) in 500 mL MilliQ water. Then, 0.2 M phosphate buffer (pH 7.0) was prepared by mixing 0.2 M sodium dihydrogenphosphate buffer and 0.2 M disodium phosphate buffer. A total of 2 g of mucin (Sigma-Aldrich Co. LLC, St. Louis, MO, USA), which was partially purified powder derived from the porcine stomach (bound sialic acid 0.5%–1.5%; free N-acetylneuraminic acid ≤0.2%), was dissolved per 100 mL of 10 mM phosphate buffer (pH 7.0). The mixture was autoclaved for 20 minutes at 121°C and centrifuged at 9,000 × g for 30 minutes at room temperature. The 60 mL of the supernatant was added to 240 mL of M9 medium.
All media were dispensed into sterile test tubes with two-position caps (Caplugs, Buffalo, NY, USA), moved to a vinyl anaerobic chamber (Coy Laboratory Products Inc., Grass Lake, MI, USA), and incubated for 24 hours to replace the medium with anaerobic conditions before the start of the experiment.

Mutated gene screening

Three groups of wild-type BALB/c GF mice (experiment 1 [male; n = 3], experiment 2 [male; n = 2], and experiment 3 [female; n = 3]) were fed a high-MAC diet in independent axenic isolators. ΔmutL and ΔmutS grown at 37°C in LB medium under aerobic conditions were diluted to 5 × 108 colony forming units per mL in phosphate-buffered saline (PBS). The animals in experiments 1 and 2 were gavaged with 200 µL of the ΔmutL suspension. The animals in experiment 3 were gavaged with 200 µL of the ΔmutS suspension. The mice were individually caged after inoculation. ΔmutL-inoculated mouse feces were collected on days 14, 28, 56, and 84. We could only sample feces of ΔmutL in experiment 2 on day 84 due to human error. After sampling ΔmutS-inoculated mouse feces on day 21, additional young BALB/c GF mice were co-housed to transfer ΔmutS to new mice from elder mice. The feces of newly inoculated ΔmutS were collected at day 84. Same as above, feces at days 21 and 84 in experiment 3 were sampled from different mouse individuals, but the microbial population was identical. All feces were stored at −80°C until use.
DNA was extracted from feces as previously described (54). Briefly, samples were incubated with 15 mg/mL lysozyme (FUJIFILM Wako Pure Chemical Corporation, Osaka, Osaka, Japan) at 37°C overnight. Then, achromopeptidase (FUJIFILM Wako Pure Chemical Corporation, Osaka, Osaka, Japan) was added to the lysates at a final concentration of 600 U/mL, and the lysates were incubated at 37°C for 8 h. After adding SDS and proteinase K (Merck KGaA, Darmstadt, Hessen, Germany) at final concentrations of 1% and 1 mg/mL, respectively, the lysates were incubated at 55°C overnight. The bacterial genomic DNA was extracted with the standard phenol/chloroform/isoamyl alcohol protocol.
After DNA extraction from fecal samples, whole genome sequencing was performed using HiSeq 2000 or NovaSeq 6000 (Illumina Inc., San Diego, CA, USA). The obtained reads were mapped against a reference sequence of E. coli K-12 MG1655 (NC_000913) using BWA version 0.7.5a (55), and duplicate sequences were eliminated using Picard version 1.97 (https://broadinstitute.github.io/picard/index.html). Reads were aligned by position using SAMtools mpileup version 0.1.18 (56), and variant calling and strand bias removal were performed using VarScan version 2.3.6 (57). Genes that carried an allele frequency of 20% or higher in a mouse were detected as mutated genes. Reads were then annotated by SnpEff version 3.4i (25), and differences from the reference sequence were extracted using in-house scripts. Mutations that were estimated to have “HIGH” or ”MODERATE” effects on protein functions by SnpEff version 3.4i were used in the following analysis.
In the comparison of mutated genes in mice fed different diets, we extracted mutated genes that were detected in more than half of the mice in the high-MAC diet-fed mice or low-MAC diet-fed mice, respectively (i.e., mutated genes detected in five or more mice in the high-MAC diet group, in which there were eight mice in total, and in more four or more mice in the low-MAC diet group, in which there were six mice in total).

In vivo competition assays between ΔgatC and multiple-mutant strains

For competition between ΔgatC and multiple-mutant strains, 8- to 12-week-old male and female BALB/c GF mice were fed with high-MAC diet in axenic isolators. Escherichia coli K-12 BW25113 ΔgatCΔintS[cmR], ∆gatCaraC[kmR], ∆gatCmalI[kmR], ∆gatCmelR[kmR], ΔgatCΔaraCΔmalI[kmR], and ∆gatCaraCmalImelR[kmR] were grown at 37°C in LB medium under aerobic conditions. They were diluted to 1 × 108 CFU/mL in PBS and mixed so that the ratio of ΔgatCΔintS[cmR] to mixtures of multiple-mutant strains was 1:1. The animals were gavaged with 200 µL of the suspension. Low-MAC diet-fed mice were gavaged after 1 week of diet acclimatization. We housed four mice in an isolator as a group, two mice housed in one cage, divided by sex (i.e., two sex-separated cages of two mice each, all within one isolator). Mouse feces were collected on days 1, 3, 5, 7, 10, and 14. A freshly collected fecal sample, weighing approximately 10–20 mg, was dissolved in LB liquid medium at a volume (µL) equivalent to the mass of the sample (mg) multiplied by 50. The original solution was diluted to 10−3, 10−5, and 10−7. Each solution was applied to 1.5% LB agar plates containing 25 µg/mL chloramphenicol for ΔgatCΔintS[cmR] and 1.5% LB agar plates containing 25 µg/mL kanamycin for multiple-mutant strains at 37°C overnight under aerobic conditions. The colonies were counted for each plate. The competition index was calculated by
CFU of multiplemutant strainCFU of ΔgatC

Metabolite extraction

Metabolites in the cecal, fecal, and colon samples were extracted as previously described (58). Briefly, samples were lyophilized in the freeze dryer (TAITEC Corporation, Koshigaya, Saitama, Japan) and disrupted with a dispensing spoon. Ten-milligram samples were combined with four 3.0 mm zirconia/silica beads and 100 mg of 0.1 mm beads (TOMY SEIKO Co. Ltd., Nerima, Tokyo, Japan), and 500 µL of methanol-dissolved metabolites was extracted by vigorous shaking at 1,500 rpm for 5 minutes using the Shake Master NEO. (Bio Medical Science Co. Ltd., Shinjuku, Tokyo, Japan). The mixture was then cleaned by further shaking at 1,500 rpm for 5 minutes with 200 µL of Milli-Q water and 500 µL of chloroform containing 20 µM methionine sulfone, 20 µM 2-(N-morpholino) ethanesulfonic acid (MES) and 20 µM D-camphor-10-sulfonic acid (CSA) as internal standards for CE-TOFMS, and 200 µM 13C6 glucose as an internal standard for LC-MS/MS. The suspension was centrifuged at 4,600 × g for 30 minutes at 4°C in high-speed, refrigerated microcentrifuges (TOMY SEIKO Co., Ltd., Nerima, Tokyo, Japan), and 300 µL of the resulting supernatant was divided into two tubes of 150 µL each for measurement using CE-TOFMS and LC-MS/MS. The supernatant was transferred to a 5-kDa-cutoff filter column (Ultrafree MC-PLHCC 250/pk for Metabolome Analysis, Human Metabolome Technologies, Inc., Tsuruoka, Yamagata, Japan) and centrifuged at 4°C at 9,100 × g overnight. Then, 30 µL of the flow-through for LC-MS/MS measurement was transferred to a screw-top vial (Agilent Technologies, Inc., Santa Clara, CA, U.S.A.), and stored in a freezer at −80°C until use. The flow-through for CE-TOFMS measurement was evaporated in the CentriVap Centrifugal Vacuum Concentrator (Labconco Corporation., Kansas, MO, U.S.A.). The evaporated samples were stored at −80°C until use. The evaporated samples were dissolved in 50 µL of 200 µM trimesate and 3-aminopyrrolidine in Milli-Q water.

Metabolome measurement using LC-MS/MS

LC-MS/MS analyses were performed as previously described (59), but with some modifications in galactitol measurement. Briefly, for galactitol measurement, a hydrophilic interaction chromatography (HILIC) polymer-based column (HILICpak VG-50 4E; 4.6 mm inner diameter [i.d.] × 250 mm length; 5 µm; Showa Denko K.K., Tokyo, Japan) was used for the separation of the sample solutions. The initial mobile phase was 93% acetonitrile and 7% Milli-Q water, and the acetonitrile gradient profile was 80%, 60%, and 93% at 30 min, 35 min, and 40 min, respectively. The flow rate was 1 mL min−1. For other sugar measurements, a HILIC amino column (Asahipak NH2P-50 4E; 4.6 mm i.d. × 250 mm length; 5 µm; Showa Denko K.K., Tokyo, Japan) was used for the separation of the sample solutions with guard column (Asahipak NH2P-50G 4A; 4.6 mm i.d. ×10 mm length; 5 µm; Showa Denko K.K., Tokyo, Japan). The initial mobile phase was 80% acetonitrile and 20% Milli-Q water, and the acetonitrile gradient profile was 70%, 60%, and 80% at 23 min, 35 min, and 40 min, respectively. The flow rate was 0.8 mL min−1.
The injection volume was 1 µL, and the duration was 40 minutes for each sample. The ion spray voltage was −4,500 V and the ion source temperature was 500°C. The peak detection and calculation of sugar concentration were performed with Analyst ver. 1.6.3 (Sciex, Framingham, MA, USA).

Metabolome measurement using CE-TOFMS

Both positive and negative modes by CE-TOFMS were used for measuring the concentration of metabolites (60, 61). An Agilent capillary electrophoresis system (Agilent Technologies, Inc., Santa Clara, CA, USA). was used in all CE-TOFMS experiments as previously described (62). Detected peak areas were normalized with methionine sulfone for cationic metabolites and CSA for anionic metabolites. The peak detection and calculation of concentration were performed with MasterHands ver. 2.19.0.1 (63).

In vivo RNA-seq of ΔgatC and ΔgatCΔaraCΔmalI

12-week-old male BALB/c GF mice (n = 4) were fed a High-MAC or Low-MAC diet in axenic isolators. Escherichia coli K-12 BW25113 ΔgatC[cmR] and ∆gatCaraCmalI[kmR] were grown at 37°C in LB medium under aerobic conditions. They were diluted to 1 × 108 CFU/mL in PBS. The animals were gavaged with 200 µL of the suspensions. After 14 days following inoculation, mice were dissected to obtain cecal contents.
RNA was extracted from the cecal contents of mice with NucleoSpin RNA Stool (Takara Bio Inc., Kusatsu, Shiga, Japan) following the manufacturer’s instructions. Then, rRNA was removed from the total RNA using Illumina Ribo-Zero Plus rRNA Depletion Kit (Illumina, Inc., San Diego, CA, USA).
RNA sequencing was performed using the Illumina NovaSeq 6000 (Illumina, Inc., San Diego, CA, USA) instrument in the paired-end 2 × 150 bp cycle mode by using NEBNext Ultra II RNA Library Prep Kit for Illumina (New England Biolabs, Inc. Ipswich, MA, USA).
The sequence generated 223,906,001 paired-end reads obtained from 12 samples. The following analysis was performed on Galaxy (https://galaxyproject.org/) (64). The quality score of the reads was visualized using FastQC (Galaxy version 0.73 + galaxy0) (65). For quality control, Trimmomatic (Galaxy version 0.38.0) (66) was performed with the following parameters: ILLUMINACLIP 2:30:10; LEADING = 20; TRAILING = 20; SLIDINGWINDOW = 4:15; MINLEN = 36. The reads that passed quality control were mapped to the Escherichia coli K-12 BW25113 reference genome (NZ_CP009273), provided by National Center for Biotechnology Information, using HISAT2 (Galaxy version 2.2.1+galaxy0) (67). Mapped reads were counted using featureCounts (Galaxy version 2.0.1+galaxy2) (68). Differentially expressed genes between ΔgatCΔaraCΔmalI and ΔgatC were calculated using DESeq2 (Galaxy version 2.11.40.7 + galaxy1) (42). Gene regulation was searched using RegulonDB version 10.10 (https://regulondb.ccg.unam.mx/index.jsp) (69)

In vitro competition assay in M9 medium

Escherichia coli K-12 BW25113 ΔgatC[cmR], ∆gatCaraC[kmR], ∆gatCmalI[kmR], ∆gatCmelR[kmR], and ∆gatCaraCmalI[kmR] were grown at 37°C in LB medium under anaerobic conditions in a vinyl anaerobic chamber (Coy Laboratory Products, Inc., Grass Lake, MI, US.) at 37°C with oxygen below 20 ppm. The anaerobic chamber’s gas composition was nitrogen, 10% carbon dioxide, and 4.85% hydrogen. Strains cultured in LB medium were washed with PBS twice and OD600 was measured with Biomate 3 (Thermo Fisher Scientific Inc., Waltham, MA, USA). The suspension was mixed so that the ratio of ΔgatC[cmR] to double or triple-mutant strains was 1:1. The mixed cultures were diluted to OD600 = 0.1 for M9 +0.4% mucin medium or OD600 = 0.01 for other M9 media. Three mL of the solution were dispensed into sterile test tubes with two-position caps (Caplugs, Buffalo, NY, USA) and incubated at 37°C under anaerobic conditions in a vinyl anaerobic chamber for 14 days. In M9 + 0.4% mucin medium, the bacterial cultures were subcultured every 48 hours into 3 mL of fresh medium at a 1/10 dilution rate. In other M9 media, the bacterial cultures were subcultured every 24 hours into 3 mL of fresh medium at a 1/100 dilution rate.
After 14 days of inoculation, the culture medium was diluted to 10−3, 10−4, and 10−5 with PBS. Fifty microliters of the diluted solutions was applied to 1.5% LB agar plates containing 25 µg/mL chloramphenicol for ΔgatC[cmR] and 1.5% LB agar plates containing 25 µg/mL kanamycin for double or triple-mutant strains at 37°C overnight under aerobic conditions. The colonies formed were counted for each plate. The competition index was calculated by
CFU of double or triplemutant strainCFU of ΔgatC

Statistical analysis

Statistical analyses were performed in R software version 4.1.2 (70). To compare the abundance of ΔgatC and multiple-mutant strains, a paired t test was performed with t.test function in the stats package version 4.1.2 and the P values were corrected with Holm’s correction for multiple comparisons with p.adjust function in the stats package. To test for differences in sugar compositions, PERMANOVA (71) was performed with adonis function in the vegan package version 2.5-7 (72). To compare the competition index of ΔgatCΔaraΔmalIΔmelR and ΔgatC between high-MAC diet-fed mice and low-MAC diet-fed mice, two-way ANOVA was performed with the aov function in the stats package. To compare metabolite concentration between high-MAC diet-fed mice and low-MAC diet-fed mice, Welch’s t test was performed and the P values were corrected with Benjamini-Hochberg correction for multiple comparisons with the p.adjust function in the stats package. Adjusted P < 0.05 were considered statistically significant.
Schematic representation of experiments was drawn with Adobe Illustrator version 26.0.1 (Adobe Inc., San Jose, CA, USA). Plots were generated with the ggplot2 package version 3.3.5 (73), the ggVennDiagram package version 1.2.0 (74), and the ComplexHeatmap package version 2.13.1 (75) in R.

ACKNOWLEDGMENTS

We thank Dr. Ryuichi Koga for providing K-12 BW25113 hyper-mutator strain (ΔmutS); Drs. Yasuyuki Ohkawa, Tetsuya Hayashi, Yasuhiro Gotoh, and Keiji Nakamura for genome sequencing; Dr. Yoshitoshi Ogura for data analysis; Dr. Kenji Nakahigashi for helpful advice; Mrs. Mitsuko Komatsu, Noriko Kagata, Maho Nakamaru, Noriko Fukuda, Maki Ohishi, Ayano Ueno, Yu Obana, and Mr. Tatsuji Takahasi for the technical support. This work was partly performed in the Cooperative Research Project Program of the Medical Institute of Bioregulation, Kyushu University.
Funding came from the following: Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Number JP21J11019 (T.T.), JSPS KAKENHI Grant Number JP22H03541 (S.F.), Japan Science and Technology Agency (JST) ERATO Grant Number JPMJER1902 (S.F.), Japan Agency of Medical Research and Development (AMED) Grant Number 23gm1010009 (S.F.), Food Science Institute Foundation (S.F.), Yamagata Prefectural Government and the City of Tsuruoka (M.T. and K.A.).
Conceptualization was done by T.T. and S.F. Resources were acquired by N.O. and H.M. Investigation was done by T.T., N.O., E.M., H.O., S.S., Y.A., M.W., and T.S. Formal analysis was done by K.A. and T.T. Supervision was done by M.T. and S.F. Writing—Original Draft was done by T.T., N.O., J.Y., and S.F. Writing—Review & Editing was done by all authors.

SUPPLEMENTAL MATERIAL

Supplemental material - msystems.01123-23-s0001.pdf
Fig. S1 to S5. Tables S1 to S5.
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.

REFERENCES

1.
Sender R, Fuchs S, Milo R. 2016. Revised estimates for the number of human and bacteria cells in the body. PLoS Biol 14:e1002533.
2.
Qin J, Li R, Raes J, Arumugam M, Burgdorf KS, Manichanh C, Nielsen T, Pons N, Levenez F, Yamada T, et al. 2010. A human gut microbial gene catalogue established by metagenomic sequencing. Nature 464:59–65.
3.
Nishijima S, Suda W, Oshima K, Kim SW, Hirose Y, Morita H, Hattori M. 2016. The gut microbiome of healthy Japanese and its microbial and functional uniqueness. DNA Res 23:125–133.
4.
Goodrich JK, Waters JL, Poole AC, Sutter JL, Koren O, Blekhman R, Beaumont M, Van Treuren W, Knight R, Bell JT, Spector TD, Clark AG, Ley RE. 2014. Human genetics shape the gut microbiome. Cell 159:789–799.
5.
Rothschild D, Weissbrod O, Barkan E, Kurilshikov A, Korem T, Zeevi D, Costea PI, Godneva A, Kalka IN, Bar N, et al. 2018. Environment dominates over host genetics in shaping human gut microbiota. Nature 555:210–215.
6.
David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Wolfe BE, Ling AV, Devlin AS, Varma Y, Fischbach MA, Biddinger SB, Dutton RJ, Turnbaugh PJ. 2014. Diet rapidly and reproducibly alters the human gut microbiome. Nature 505:559–563.
7.
Maier TV, Lucio M, Lee LH, VerBerkmoes NC, Brislawn CJ, Bernhardt J, Lamendella R, McDermott JE, Bergeron N, Heinzmann SS, Morton JT, González A, Ackermann G, Knight R, Riedel K, Krauss RM, Schmitt-Kopplin P, Jansson JK. 2017. Impact of dietary resistant starch on the human gut microbiome, metaproteome, and metabolome. mBio 8:e01343-17.
8.
Desai MS, Seekatz AM, Koropatkin NM, Kamada N, Hickey CA, Wolter M, Pudlo NA, Kitamoto S, Terrapon N, Muller A, Young VB, Henrissat B, Wilmes P, Stappenbeck TS, Núñez G, Martens EC. 2016. A dietary fiber-deprived gut microbiota degrades the colonic mucus barrier and enhances pathogen susceptibility. Cell 167:1339–1353.
9.
Sonnenburg ED, Smits SA, Tikhonov M, Higginbottom SK, Wingreen NS, Sonnenburg JL. 2016. Diet-induced extinctions in the gut microbiota compound over generations. Nature 529:212–215.
10.
Yilmaz B, Mooser C, Keller I, Li H, Zimmermann J, Bosshard L, Fuhrer T, Gomez de Agüero M, Trigo NF, Tschanz-Lischer H, Limenitakis JP, Hardt W-D, McCoy KD, Stecher B, Excoffier L, Sauer U, Ganal-Vonarburg SC, Macpherson AJ. 2021. Long-term evolution and short-term adaptation of microbiota strains and sub-strains in mice. Cell Host Microbe 29:650–663.
11.
Shepherd ES, DeLoache WC, Pruss KM, Whitaker WR, Sonnenburg JL. 2018. An exclusive metabolic niche enables strain engraftment in the gut microbiota. Nature 557:434–438.
12.
Odamaki T, Kato K, Sugahara H, Hashikura N, Takahashi S, Xiao J-Z, Abe F, Osawa R. 2016. Age-related changes in gut microbiota composition from newborn to centenarian: a cross-sectional study. BMC Microbiol 16:90.
13.
Palmer C, Bik EM, DiGiulio DB, Relman DA, Brown PO, Ruan Y. 2007. Development of the human infant intestinal microbiota. PLoS Biol 5:e177.
14.
Bäckhed F, Roswall J, Peng Y, Feng Q, Jia H, Kovatcheva-Datchary P, Li Y, Xia Y, Xie H, Zhong H, et al. 2015. Dynamics and stabilization of the human gut microbiome during the first year of life. Cell Host & Microbe 17:690–703.
15.
Crook N, Ferreiro A, Gasparrini AJ, Pesesky MW, Gibson MK, Wang B, Sun X, Condiotte Z, Dobrowolski S, Peterson D, Dantas G. 2019. Adaptive strategies of the candidate probiotic E. coli nissle in the mammalian gut. Cell Host Microbe 25:499–512.
16.
Barroso-Batista J, Sousa A, Lourenço M, Bergman M-L, Sobral D, Demengeot J, Xavier KB, Gordo I. 2014. The first steps of adaptation of Escherichia coli to the gut are dominated by soft sweeps. PLoS Genet 10:e1004182.
17.
Barroso-Batista J, Pedro MF, Sales-Dias J, Pinto CJG, Thompson JA, Pereira H, Demengeot J, Gordo I, Xavier KB. 2020. Specific Eco-evolutionary contexts in the mouse gut reveal Escherichia coli metabolic versatility. Curr Biol 30:1049–1062.
18.
Lescat M, Launay A, Ghalayini M, Magnan M, Glodt J, Pintard C, Dion S, Denamur E, Tenaillon O. 2017. Using long-term experimental evolution to uncover the patterns and determinants of molecular evolution of an Escherichia coli natural isolate in the streptomycin-treated mouse gut. Mol Ecol 26:1802–1817.
19.
Dapa T, Ramiro RS, Pedro MF, Gordo I, Xavier KB. 2022. Diet leaves a genetic signature in a keystone member of the gut microbiota. Cell Host Microbe 30:183–199.
20.
Otsuka Y, Muto A, Takeuchi R, Okada C, Ishikawa M, Nakamura K, Yamamoto N, Dose H, Nakahigashi K, Tanishima S, Suharnan S, Nomura W, Nakayashiki T, Aref WG, Bochner BR, Conway T, Gribskov M, Kihara D, Rudd KE, Tohsato Y, Wanner BL, Mori H. 2015. GenoBase: comprehensive resource database of Escherichia coli K-12. Nucleic Acids Res 43:D606–17.
21.
Koga R, Moriyama M, Onodera-Tanifuji N, Ishii Y, Takai H, Mizutani M, Oguchi K, Okura R, Suzuki S, Gotoh Y, Hayashi T, Seki M, Suzuki Y, Nishide Y, Hosokawa T, Wakamoto Y, Furusawa C, Fukatsu T. 2022. Single mutation makes Escherichia coli an insect mutualist. Nat Microbiol 7:1141–1150.
22.
Lee H, Popodi E, Tang H, Foster PL. 2012. Rate and molecular spectrum of spontaneous mutations in the bacterium Escherichia coli as determined by whole-genome sequencing. Proc Natl Acad Sci U S A 109:E2774–83.
23.
Giraud A, Matic I, Tenaillon O, Clara A, Radman M, Fons M, Taddei F. 2001. Costs and benefits of high mutation rates: adaptive evolution of bacteria in the mouse gut. Science 291:2606–2608.
24.
Kang M, Kim K, Choe D, Cho S, Kim SC, Palsson B, Cho BK. 2019. Inactivation of a mismatch-repair system diversifies genotypic landscape of Escherichia coli during adaptive laboratory evolution. Front Microbiol 10:1845.
25.
Cingolani P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, Land SJ, Lu X, Ruden DM. 2012. A program for annotating and predicting the effects of single nucleotide polymorphisms, Snpeff: SNPs in the genome of drosophila melanogaster strain w1118; iso-2; iso-3. Fly 6:80–92.
26.
Nobelmann B, Lengeler JW. 1996. Molecular analysis of the gat genes from Escherichia coli and of their roles in galactitol transport and metabolism. J Bacteriol 178:6790–6795.
27.
Miyada CG, Stoltzfus L, Wilcox G. 1984. Regulation of the araC gene of Escherichia coli: catabolite repression, autoregulation, and effect on araBAD expression. Proc Natl Acad Sci U S A 81:4120–4124.
28.
Gallegos MT, Schleif R, Bairoch A, Hofmann K, Ramos JL. 1997. Arac/XylS family of transcriptional regulators. Microbiol Mol Biol Rev 61:393–410.
29.
Reidl J, Römisch K, Ehrmann M, Boos W. 1989. Mali, a novel protein involved in regulation of the maltose system of Escherichia coli, is highly homologous to the repressor proteins GalR, CytR, and LacI. J Bacteriol 171:4888–4899.
30.
Barroso-Batista J, Demengeot J, Gordo I. 2015. Adaptive immunity increases the pace and predictability of evolutionary change in commensal gut bacteria. Nat Commun 6:8945.
31.
Sousa A, Ramiro RS, Barroso-Batista J, Güleresi D, Lourenço M, Gordo I. 2017. Recurrent reverse evolution maintains polymorphism after strong bottlenecks in commensal gut bacteria. Mol Biol Evol 34:2879–2892.
32.
Vasquez KS, Willis L, Cira NJ, Ng KM, Pedro MF, Aranda-Díaz A, Rajendram M, Yu FB, Higginbottom SK, Neff N, Sherlock G, Xavier KB, Quake SR, Sonnenburg JL, Good BH, Huang KC. 2021. Quantifying rapid bacterial evolution and transmission within the mouse intestine. Cell Host & Microbe 29:1454–1468.
33.
Stringer AM, Currenti S, Bonocora RP, Baranowski C, Petrone BL, Palumbo MJ, Reilly AA, Zhang Z, Erill I, Wade JT. 2014. Genome-scale analyses of Escherichia coli and Salmonella enterica AraC reveal noncanonical targets and an expanded core regulon. J Bacteriol 196:660–671.
34.
Macpherson AJ, Jones-Mortimer MC, Horne P, Henderson PJ. 1983. Identification of the GalP galactose transport protein of Escherichia coli. J Biol Chem 258:4390–4396.
35.
Holmes RP, Russell RR. 1972. Mutations affecting amino sugar metabolism in Escherichia coli K-12. J Bacteriol 111:290–291.
36.
Peri KG, Goldie H, Waygood EB. 1990. Cloning and characterization of the N-acetylglucosamine operon of Escherichia coli. Biochem Cell Biol 68:123–137.
37.
Berteau O, Guillot A, Benjdia A, Rabot S. 2006. A new type of bacterial sulfatase reveals a novel maturation pathway in prokaryotes. J Biol Chem 281:22464–22470.
38.
Schirmer A, Kolter R. 1998. Computational analysis of bacterial sulfatases and their modifying enzymes. Chem Biol 5:R181–6.
39.
Benjdia A, Martens EC, Gordon JI, Berteau O. 2011. Sulfatases and a radical S-adenosyl-L-methionine (AdoMet) enzyme are key for mucosal foraging and fitness of the prominent human gut symbiont, Bacteroides thetaiotaomicron. J Biol Chem 286:25973–25982.
40.
Humbert R, Simoni RD. 1980. Genetic and biomedical studies demonstrating a second gene coding for asparagine synthetase in Escherichia coli. J Bacteriol 142:212–220.
41.
Schwartz JH, Reeves JY, Broome JD. 1966. Two L-asparaginases from E. coli and their action against tumors. Proc Natl Acad Sci U S A 56:1516–1519.
42.
Love MI, Huber W, Anders S. 2014. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15:550.
43.
Everard A, Belzer C, Geurts L, Ouwerkerk JP, Druart C, Bindels LB, Guiot Y, Derrien M, Muccioli GG, Delzenne NM, de Vos WM, Cani PD. 2013. Cross-talk between Akkermansia muciniphila and intestinal epithelium controls diet-induced obesity. Proc Natl Acad Sci U S A 110:9066–9071.
44.
Sonnenburg JL, Xu J, Leip DD, Chen CH, Westover BP, Weatherford J, Buhler JD, Gordon JI. 2005. Glycan foraging in vivo by an intestine-adapted bacterial symbiont. Science 307:1955–1959.
45.
Derrien M, Vaughan EE, Plugge CM, de Vos WM. 2004. Horizontal gene transfer overrides mutation in Escherichia coli colonizing the mammalian gut. Int J Syst Evol Microbiol 54:1469–1476.
46.
Tramontano M, Andrejev S, Pruteanu M, Klünemann M, Kuhn M, Galardini M, Jouhten P, Zelezniak A, Zeller G, Bork P, Typas A, Patil KR. 2018. Nutritional preferences of human gut bacteria reveal their metabolic idiosyncrasies. Nat Microbiol 3:514–522.
47.
Frazão N, Sousa A, Lässig M, Gordo I. 2019. Horizontal gene transfer overrides mutation in Escherichia coli colonizing the mammalian gut. Proc Natl Acad Sci U S A 116:17906–17915.
48.
Frazão N, Konrad A, Amicone M, Seixas E, Güleresi D, Lässig M, Gordo I. 2022. Two modes of evolution shape bacterial strain diversity in the mammalian gut for thousands of generations. Nat Commun 13:5604.
49.
Smith HW. 1975. Survival of orally administered E. coli K 12 in alimentary tract of man. Nature 255:500–502.
50.
Jang J, Hur HG, Sadowsky MJ, Byappanahalli MN, Yan T, Ishii S. 2017. Environmental Escherichia coli: ecology and public health implications-a review. J Appl Microbiol 123:570–581.
51.
Rossen NG, Fuentes S, van der Spek MJ, Tijssen JG, Hartman JHA, Duflou A, Löwenberg M, van den Brink GR, Mathus-Vliegen EMH, de Vos WM, Zoetendal EG, D’Haens GR, Ponsioen CY. 2015. Findings from a randomized controlled trial of fecal transplantation for patients with ulcerative colitis. Gastroenterology 149:110–118.
52.
Markowiak P, Śliżewska K. 2017. Effects of probiotics, prebiotics, and synbiotics on human health. Nutrients 9:1021.
53.
Baba T, Ara T, Hasegawa M, Takai Y, Okumura Y, Baba M, Datsenko KA, Tomita M, Wanner BL, Mori H. 2006. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol Syst Biol 2:0008.
54.
Yang J, Tsukimi T, Yoshikawa M, Suzuki K, Takeda T, Tomita M, Fukuda S, Gilbert JA. 2019. Cutibacterium acnes (Propionibacterium acnes) 16S rRNA genotyping of microbial samples from possessions contributes to owner identification. mSystems 4.
55.
Li H, Durbin R. 2009. Fast and accurate short read alignment with burrows-wheeler transform. Bioinformatics 25:1754–1760.
56.
Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R, Genome Project Data Processing Subgroup. 2009. The sequence alignment/map format and Samtools. Bioinformatics 25:2078–2079.
57.
Koboldt DC, Chen K, Wylie T, Larson DE, McLellan MD, Mardis ER, Weinstock GM, Wilson RK, Ding L. 2009. VarScan: variant detection in massively parallel sequencing of individual and pooled samples. Bioinformatics 25:2283–2285.
58.
Yamamoto Y, Nakanishi Y, Murakami S, Aw W, Tsukimi T, Nozu R, Ueno M, Hioki K, Nakahigashi K, Hirayama A, Sugimoto M, Soga T, Ito M, Tomita M, Fukuda S. 2018. A metabolomic-based evaluation of the role of commensal microbiota throughout the gastrointestinal tract in mice. Microorganisms 6:101.
59.
Ogura T, Wakayama M, Ashino Y, Kadowaki R, Sato M, Soga T, Tomita M. 2020. Effects of feed crops and boiling on chicken egg yolk and white determined by a metabolome analysis. Food Chem 327:127077.
60.
Soga T, Ohashi Y, Ueno Y, Naraoka H, Tomita M, Nishioka T. 2003. Quantitative metabolome analysis using capillary electrophoresis mass spectrometry. J Proteome Res 2:488–494.
61.
Wakayama M, Hirayama A, Soga T. 2015. Capillary electrophoresis-mass spectrometry. Methods Mol Biol 1277:113–122.
62.
Ishii C, Nakanishi Y, Murakami S, Nozu R, Ueno M, Hioki K, Aw W, Hirayama A, Soga T, Ito M, Tomita M, Fukuda S. 2018. A metabologenomic approach reveals changes in the intestinal environment of mice fed on American diet. Int J Mol Sci 19:4079.
63.
Sugimoto M, Hirayama A, Robert M, Abe S, Soga T, Tomita M. 2010. Prediction of metabolite identity from accurate mass, migration time prediction and isotopic pattern information in CE-TOFMS data. Electrophoresis 31:2311–2318.
64.
Afgan E, Baker D, Batut B, van den Beek M, Bouvier D, Cech M, Chilton J, Clements D, Coraor N, Grüning BA, Guerler A, Hillman-Jackson J, Hiltemann S, Jalili V, Rasche H, Soranzo N, Goecks J, Taylor J, Nekrutenko A, Blankenberg D. 2018. The galaxy platform for accessible, reproducible and collaborative biomedical analyses. Nucleic Acids Res 46:W537–W544.
65.
Andrews S. 2010. FastQC: a quality control tool for high throughput sequence data. Available from: http://www.bioinformatics.babraham.ac.uk/projects/fastqc/
66.
Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for illumina sequence data. Bioinformatics 30:2114–2120.
67.
Kim D, Langmead B, Salzberg SL. 2015. HISAT: a fast spliced aligner with low memory requirements. Nat Methods 12:357–360.
68.
Liao Y, Smyth GK, Shi W. 2014. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30:923–930.
69.
Santos-Zavaleta A, Salgado H, Gama-Castro S, Sánchez-Pérez M, Gómez-Romero L, Ledezma-Tejeida D, García-Sotelo JS, Alquicira-Hernández K, Muñiz-Rascado LJ, Peña-Loredo P, Ishida-Gutiérrez C, Velázquez-Ramírez DA, Del Moral-Chávez V, Bonavides-Martínez C, Méndez-Cruz C-F, Galagan J, Collado-Vides J. 2019. RegulonDB V 10.5: tackling challenges to unify classic and high throughput knowledge of gene regulation in E. coli K-12. Nucleic Acids Res 47:D212–D220.
70.
Ihaka R, Gentleman R. 1996. R: a language for data analysis and graphics. J Comput and Graphical Stat 5:299–314.
71.
Anderson MJ. 2001. A new method for non-parametric multivariate analysis of variance. Austral Ecology 26:32–46.
72.
Li S, Chen K, Vähänissi V, Radevici I, Savin H, Oksanen J. 2022. Electron injection in metal assisted chemical etching as a fundamental mechanism for electroless electricity generation. J Phys Chem Lett 13:5648–5653.
73.
Wickham H. 2016. Ggplot2. In Ggplot2: elegant graphics for data analysis. Springer-Verlag New York, Cham.
74.
Gao C-H, Yu G, Cai P. 2021. ggVennDiagram: an intuitive, easy-to-use, and highly customizable R package to generate venn diagram. Front Genet 12:706907.
75.
Gu Z, Eils R, Schlesner M. 2016. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32:2847–2849.

Information & Contributors

Information

Published In

cover image mSystems
mSystems
Volume 9Number 220 February 2024
eLocator: e01123-23
Editor: Suzanne Lynn Ishaq, The University of Maine, Orono, Maine, USA
PubMed: 38205998

History

Received: 23 October 2023
Accepted: 15 November 2023
Published online: 11 January 2024

Keywords

  1. gut microbiota
  2. Escherichia coli
  3. genetic mutation
  4. intestinal colonization
  5. intestinal nutrient

Data Availability

Whole genome sequencing and in vivo RNA-seq data have been deposited in the DNA Data Bank of Japan (DDBJ) Sequence Read Archive (http://trace.ddbj.nig.ac.jp/dra/). The accession numbers for the mutated gene screening are DRA015347, DRA016043, DRA016044, and DRA016621. The accession number for in vivo RNA-seq is DRA015348. The original code has been deposited at https://github.com/t-tsukimi/ecoli_mutation. Any additional information required to reanalyze the data reported in this paper is available from the corresponding author upon request.

Contributors

Authors

Institute for Advanced Biosciences, Keio University, Tsuruoka, Japan
Systems Biology Program, Graduate School of Media and Governance, Keio University, Fujisawa, Japan
Author Contributions: Conceptualization, Formal analysis, Investigation, Writing – original draft, and Writing – review and editing.
Transborder Medical Research Center, Institute of Medicine, University of Tsukuba, Tsukuba, Japan
Author Contributions: Investigation, Resources, Writing – original draft, and Writing – review and editing.
Suguru Shigemori
Transborder Medical Research Center, Institute of Medicine, University of Tsukuba, Tsukuba, Japan
Author Contributions: Investigation and Writing – review and editing.
Present address: Institute for Biomedical Sciences, Shinshu University, Kami-ina, Japan
Institute for Advanced Biosciences, Keio University, Tsuruoka, Japan
Systems Biology Program, Graduate School of Media and Governance, Keio University, Fujisawa, Japan
Faculty of Environment and Information Studies, Keio University, Fujisawa, Japan
Author Contributions: Formal analysis and Writing – review and editing.
Eiji Miyauchi
RIKEN Center for Integrative Medical Sciences, Yokohama, Japan
Institute for Molecular and Cellular Regulation, Gunma University, Maebashi, Japan
Author Contributions: Investigation and Writing – review and editing.
Institute for Advanced Biosciences, Keio University, Tsuruoka, Japan
Systems Biology Program, Graduate School of Media and Governance, Keio University, Fujisawa, Japan
Author Contributions: Writing – original draft and Writing – review and editing.
Isaiah Song
Institute for Advanced Biosciences, Keio University, Tsuruoka, Japan
Systems Biology Program, Graduate School of Media and Governance, Keio University, Fujisawa, Japan
Author Contribution: Writing – review and editing.
Yujin Ashino
Institute for Advanced Biosciences, Keio University, Tsuruoka, Japan
Author Contributions: Investigation and Writing – review and editing.
Masataka Wakayama
Institute for Advanced Biosciences, Keio University, Tsuruoka, Japan
Systems Biology Program, Graduate School of Media and Governance, Keio University, Fujisawa, Japan
Author Contributions: Investigation and Writing – review and editing.
Present address: Integrated Medical and Agricultural School of Public Health, Ehime University, Toon, Japan
Tomoyoshi Soga
Institute for Advanced Biosciences, Keio University, Tsuruoka, Japan
Systems Biology Program, Graduate School of Media and Governance, Keio University, Fujisawa, Japan
Faculty of Environment and Information Studies, Keio University, Fujisawa, Japan
Author Contributions: Investigation and Writing – review and editing.
Masaru Tomita
Institute for Advanced Biosciences, Keio University, Tsuruoka, Japan
Systems Biology Program, Graduate School of Media and Governance, Keio University, Fujisawa, Japan
Faculty of Environment and Information Studies, Keio University, Fujisawa, Japan
Author Contributions: Supervision and Writing – review and editing.
Hiroshi Ohno
RIKEN Center for Integrative Medical Sciences, Yokohama, Japan
Author Contributions: Investigation and Writing – review and editing.
Hirotada Mori
Graduate School of Biological Science, Nara Institute of Science and Technology, Ikoma, Japan
Institute of Animal Sciences, Guangdong Academy of Agricultural Sciences, Guangzhou, Guangdong, China
Author Contributions: Resources and Writing – review and editing.
Institute for Advanced Biosciences, Keio University, Tsuruoka, Japan
Systems Biology Program, Graduate School of Media and Governance, Keio University, Fujisawa, Japan
Transborder Medical Research Center, Institute of Medicine, University of Tsukuba, Tsukuba, Japan
Gut Environmental Design Group, Kanagawa Institute of Industrial Science and Technology, Kawasaki, Japan
Laboratory for Regenerative Microbiology, Juntendo University Graduate School of Medicine, Tokyo, Japan
Author Contributions: Conceptualization, Supervision, Writing – original draft, and Writing – review and editing.

Editor

Suzanne Lynn Ishaq
Editor
The University of Maine, Orono, Maine, USA

Notes

S.F. is the founder and CEO of Metagen Inc., a company dedicated to improving human health through gut environmental design. This organization had no control over the interpretation, writing, or publication of this work. Keio University is managing the terms of these arrangements according to its conflict of interest policies. All other authors declare they have no competing interests.

Metrics & Citations

Metrics

Note:

  • For recently published articles, the TOTAL download count will appear as zero until a new month starts.
  • There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
  • Citation counts come from the Crossref Cited by service.

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

View Options

Figures and Media

Figures

Media

Tables

Share

Share

Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy